TW490859B - Methods of reading and writing data from/on semiconductor memory device, and method for driving the device - Google Patents

Methods of reading and writing data from/on semiconductor memory device, and method for driving the device Download PDF

Info

Publication number
TW490859B
TW490859B TW090105477A TW90105477A TW490859B TW 490859 B TW490859 B TW 490859B TW 090105477 A TW090105477 A TW 090105477A TW 90105477 A TW90105477 A TW 90105477A TW 490859 B TW490859 B TW 490859B
Authority
TW
Taiwan
Prior art keywords
voltage
electrode
effect transistor
field effect
electric field
Prior art date
Application number
TW090105477A
Other languages
Chinese (zh)
Inventor
Yoshihisa Kato
Original Assignee
Matsushita Electric Ind Co Ltd
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by Matsushita Electric Ind Co Ltd filed Critical Matsushita Electric Ind Co Ltd
Application granted granted Critical
Publication of TW490859B publication Critical patent/TW490859B/en

Links

Classifications

    • GPHYSICS
    • G11INFORMATION STORAGE
    • G11CSTATIC STORES
    • G11C11/00Digital stores characterised by the use of particular electric or magnetic storage elements; Storage elements therefor
    • G11C11/21Digital stores characterised by the use of particular electric or magnetic storage elements; Storage elements therefor using electric elements
    • G11C11/22Digital stores characterised by the use of particular electric or magnetic storage elements; Storage elements therefor using electric elements using ferroelectric elements

Landscapes

  • Engineering & Computer Science (AREA)
  • Computer Hardware Design (AREA)
  • Non-Volatile Memory (AREA)
  • Dram (AREA)
  • Semiconductor Memories (AREA)
  • Read Only Memory (AREA)

Abstract

In a semiconductor memory device in which a ferroelectric capacitor is connected to the gate of a field effect transistor (FET), a gate charge at the threshold voltage (Vti) of the FET is represented as Qti. In a polarization-voltage characteristic exhibited by the ferroelectric capacitor where a voltage applied thereto starts to be increased on the supposition that a polarization of 0 C/cm<SP>2</SP> initially exists in the capacitor, a voltage, associated with a polarization value corresponding to Qti, is represented as Vtf. In a read operation, the intersection between the gate charge-gate voltage characteristic of the FET and the polarization-voltage characteristic of the ferroelectric capacitor is the operating point where, a worst-case polarization of 0 C/cm<SP>2</SP> exists in the capacitor after data has been retained in the capacitor. By applying a voltage Vtf+Vth to the control electrode, the data can be read out correctly until the polarization decreases to reach 0 C/cm<SP>2</SP>.

Description

490859 __案號90105477_年月曰 絛正 五、發明說明(1) [發明背景] 本發明係有關於一種將鐵電體電容串連連接於場效應 電晶體(FET)閘極之不揮發性記憶元件。 鐵電體FET (FeFET)係由鐵電體與FET之不揮發性記 憶元件組合而成。該元件之構造大致上可區分成MFS型490859 __Case No. 90105477_Yan Zhengwu 5. Description of the invention (1) [Background of the invention] The present invention relates to a non-volatile method of connecting a ferroelectric capacitor in series to a field effect transistor (FET) gate. Sexual memory element. A ferroelectric FET (FeFET) is a combination of a ferroelectric and a nonvolatile memory element of a FET. The structure of this element can be roughly divided into MFS type

FeFET、MFIS 型 FeFET、MFMIS 型 FeFET °MFS 型係指 Metal (金屬)/ Ferroelectric (鐵電體)/ Semiconductor (半導體)°MFIS 型係指 Metal (金屬)/ Ferroelectric (鐵電體)/ Insulator (絕緣體)/ Semiconductor (半FeFET, MFIS type FeFET, MFMIS type FeFET ° MFS refers to Metal (Metal) / Ferroelectric (Ferroelectric) / Semiconductor (semiconductor) ° MFIS refers to Metal (Metal) / Ferroelectric (Ferroelectric) / Insulator (Insulator ) / Semiconductor (half

導體)。另外,MFMIS型FeFET係指Metal (金屬)/conductor). In addition, MFMIS-type FeFET means Metal /

Ferroelectric (鐵電體)/ Me ta 1 (金屬)/ I nsu i a tor (絕緣體)/ Semiconductor (半導體)。 圖9係表示有關於MFMIS型FeFET構成之剖面構造圖。 如圖9所示,於P型矽(以下,以si表示)基板31上, MFMIS型FeFET含有成為汲極電極之高濃度n型植入區域 32、與成為源極電極之高濃度n型植入區域33,藉由形成 絕緣體3 4、漂移閘極電壓3 5、鐵電體3 6、控制電極3 7等所 構成。於FET之漂移閘極電壓35上,形成鐵電體電容,而 鐵電體電容之電極與FET之閘極電極是共同的。同時,以Ferroelectric (Metal) / Me ta 1 (Metal) / I nsu i a tor (Insulator) / Semiconductor (Semiconductor). FIG. 9 is a cross-sectional structure diagram showing the structure of an MFMIS-type FeFET. As shown in FIG. 9, on a P-type silicon (hereinafter, referred to as si) substrate 31, the MFMIS-type FeFET includes a high-concentration n-type implanted region 32 serving as a drain electrode and a high-concentration n-type implant serving as a source electrode. The entrance region 33 is formed by forming an insulator 34, a drift gate voltage 35, a ferroelectric 36, a control electrode 37, and the like. A ferroelectric capacitor is formed on the drift gate voltage 35 of the FET, and the electrode of the ferroelectric capacitor and the gate electrode of the FET are common. Meanwhile, with

配線連接鐵電體電容之電極與FET之閘極電極之構造也是 MFMIS 型FeFET 之一種。 接著’針對於做為MFMIS型FeFET不揮發性記憶元件之 動作加以說明。 改變控制閘極電壓(Vg)值之正負號時,如此一來,The structure that connects the electrode of the ferroelectric capacitor with the gate electrode of the FET is also a type of MFMIS FeFET. Next, the operation of the non-volatile memory element as an MFMIS-type FeFET will be described. When changing the sign of the control gate voltage (Vg) value, in this way,

490859490859

反轉。於此,定義將正電壓施加 ’而將負電壓施加於Vg以資料 鐵電體之偏極化方向因而 於Vg以資料「i」寫入 「0」寫入。 即使將控制電極浮置(floating),偏極化值仍被保 持(殘留偏極化)。資料讀出時,保持將控制電極浮置’、 為是由於電阻成分而引起漏電,藉由與配線之電容耦 &amp;而=接近接地電位之電位)t狀態,於源極/ 電藉由鐵電體電容之殘留偏極化,fet之 或負其中之一。若漂移電極電位為 ====:=成”通之狀態使 :ϊί值=間的電流值⑽)與預=㈡ 料「1」,若Ids &lt; f,則定義為資 已寫入的資料正確地讀出。J疋義為貧料「0」’便能將 (8ΓΒΓτ!〇7&quot; j ^ ^ ^ ^Reverse. Here, it is defined that a positive voltage is applied to Vg and a negative voltage is applied to Vg to write the polarization direction of the ferroelectric, and thus Vg is written to the data "i" and "0". Even if the control electrode is floated, the polarization value is maintained (residual polarization). When the data is read, keep the control electrode floating ', because the leakage caused by the resistance component, the capacitive coupling with the wiring &amp; and the potential close to the ground potential) t state, the source / electricity by iron Residual polarization of the capacitor, one of fet or negative. If the potential of the drift electrode is =====: = 成, the state of the current is: ϊί 值 = 间 的 电流 值 ⑽) and preliminary = ㈡1, if Ids &lt; f, it is defined as written The information was read correctly. J 疋 yi is lean material "0" ', then (8ΓΒΓτ! 〇7 &quot; j ^ ^ ^ ^

UrB^TaA)、而對於絕緣體 擬該等MFMIS型FeFET之動作狀能。一化矽(Sl〇2)以模 圖10係顯示FeFET之擎%千 MFMIS型FeFET之動作進專效電路圖。利用此圖1〇針對 容、42為FET、43為控制雷\ 。於圖1〇,41為鐵電體電 電極。 電極、44為源極電極、45為汲極 各參數分別設定為 srBi2Ta2〇9之厚度為2〇〇龍、介電 490859UrB ^ TaA), and for the insulator, the operation performance of these MFMIS-type FeFETs is proposed. Silicon silicon (S102) is shown in Figure 10. It is a special circuit diagram showing the operation of FeFET's MFMIS type FeFET. Use this figure 10 for capacitance, 42 for FET, and 43 for control mine. In Fig. 10, 41 is a ferroelectric electrode. Electrode, 44 is the source electrode, 45 is the drain electrode, each parameter is set to srBi2Ta2, the thickness is 200 dragons, the dielectric is 490859

瘩數為300、矯頑電壓(coercive V〇ltage)*〇8v n〇 2ii3.5nm、介電常數為3.9。另外,mfmis構造_2區 之鐵電體電容之偏極化為Pf、電極間電壓為Vf、控制電 之電壓為Vg、於MFMIS構造MIS區之FET之閘極中所工儲存的° 電荷為Qi、汲極電極之電壓為Vd、漂移閘極電極之電子斤_ 極電極之電壓為^ = ov、基板電位為、切斷裁 電體電容而使得MIS構造之FET單獨地動作時之臨限 為Vti = 0·5V 〇 0义’电i值 如此之記憶單元,以下之二式是成立的The unitary number was 300, the coercive voltage * 〇8v no 2ii 3.5 nm, and the dielectric constant was 3.9. In addition, the polarization of the ferroelectric capacitor in the mfmis structure_2 region is Pf, the voltage between the electrodes is Vf, the control voltage is Vg, and the ° charge stored in the gate of the FET in the MIS region of the MFMIS structure is Qi, the voltage of the drain electrode is Vd, the voltage of the drift gate electrode is _ = the voltage of the electrode is ^ = ov, the potential of the substrate is, the cut-off capacitor is cut off, and the threshold of the MIS structure FET is alone For Vti = 0 · 5V 〇0 meaning 'memory unit with electric value such that the following two formulas are true

Pf (Vf) - Qi (Vi) (l) yg = Vf + Vi (2) 整理此二式,Pf (Vf)-Qi (Vi) (l) yg = Vf + Vi (2)

Pf (Vf) = Qi (Vg - Vf) (3) 成立。 圖11 A係鐵電體電容之Pf對Vf特性(稱為 、 路),以下側之曲線1表示增加電壓而得到之pf值,、 之曲線2表示減少電壓而得到之”值為。pf對^特性以上側 =〇 C/cm2時的電壓,稱為抗電壓(Vc)。 、# iPf 圖11 β係以曲線51表示FET之Qi對Vi特性。 以眾所周知之M0S電容的計算方法可容易地求得。寺性為 將此Qi對Vi特性,對於Qi軸進行對稱摔作于。 向電壓軸方向偏移Vg,錢得到Qiff (Vg^ ’並且若僅 將Pf對Vf特性與Qi對(Vg _ Vf)特性重疊地 特性。 圖面上而求得其交點。該交點表示FeFET之動:二於一個Pf (Vf) = Qi (Vg-Vf) (3) holds. Figure 11 Pf vs. Vf characteristics of A-series ferroelectric capacitors (referred to as, circuit). The curve 1 on the lower side represents the value of pf obtained by increasing the voltage, and the curve 2 represents the value obtained by decreasing the voltage. ^ The voltage above the characteristic = 〇C / cm2 is called the withstand voltage (Vc). # IPf Figure 11 β represents the Qi vs. Vi characteristics of the FET by curve 51. The well-known calculation method of M0S capacitor can be easily Calculate. The characteristic is to apply this Qi to the Vi characteristic, and perform a symmetrical fall on the Qi axis. By shifting Vg toward the voltage axis, Qian gets Qiff (Vg ^ ', and if only Pf vs. Vf vs. Qi vs. (Vg _ Vf) characteristics overlap characteristics. Find the intersection point on the drawing. This intersection point indicates the movement of FeFET: two to one

第8頁 490859 _案號 90105477 五、發明說明(4) (3 ) 即可明顯地看出 Vx , Vi = Vg - Vx 〇 —月 日倏正__ 交點之蕾陶、 每若設定為Vx,則Vf二 以下,利用藉著該圖之動Page 8 490859 _Case No. 90105477 V. Explanation of the invention (4) (3) It can be seen clearly that Vx, Vi = Vg-Vx 〇— 月 日 倏 正 __ Lei Tao at the intersection, whenever Vx is set, Then Vf below two, using the movement through the graph

FeFET之動作狀態。 ^解析手法,以模擬 最初’資料「1」寫入動作 上施加正電壓(Vg &gt; 〇v)時之’亦即進行於控制電極 線52表示施加Pf對Vf特性1,2 $作點解析。圖12A係以曲 (Vg - Vf)特性。由圖12A,若结8二15V之電壓時之Qi對 可以得到Vx = 3V,於是Vf = 3v貝出父點53之電壓值,則 12V成立。 ,由式(2) ’於是Vi二 接著,資料「0」寫入動作,女曰 上施加負電壓(Vg &lt; 〇ν)時之,進行於控制電極FeFET operation status. ^ Analytical method to simulate the initial "data" 1 "write operation when a positive voltage (Vg &gt; 〇v) is applied, that is, the control electrode line 52 indicates that the application of Pf to Vf characteristics 1, 2 $ for point analysis . Fig. 12A is a curve (Vg-Vf) characteristic. From FIG. 12A, if the Qi pair at a voltage of 8 to 15V can obtain Vx = 3V, then Vf = 3v, and the voltage value of the parent point 53 is 12V. From the formula (2) ′, then Vi 2 Next, the data “0” is written, and when a negative voltage (Vg &lt; 〇ν) is applied to the data, it is performed on the control electrode.

Pf對m寺性i,2州= _15df;解析。圖12B係施加 特性' 由瞧,若讀出交點55之電壓 二Pf for m temple i, 2 states = _15df; analysis. Figure 12B shows the applied characteristics.

Vx = -3· 5V,於是Vf = -3 5V,± -H.5V成立。 由式⑴,於是Vi = 態 間 此 Vx 保存此種狀態下之資料時,控制電極雖然為浮置狀 ,但,於鐵電體與FET之漏電導致電壓降低,及各電極 電各輕合,而可s忍為控制電極電壓幾近於〇V。因 ^控制電極於浮置狀態下,假設Vg = ov (vi = Vg _ 与—Vx) 而進行討論。 於圖1 3係顯示於資料保持狀態之動作狀態之”對” 性',2及Qi對(Vg — Vf)特性56。上述之二值資料寫入、 之後’遵循鐵電體之電滞後特性而決定動作點。亦即,由 MM, Q〇l 05477 五、發明說明(5) 資料 「1 , 曰 修正 」 寫入動作上之钟+ 2作中之Vf減少之故,電、帶:口體之施加電壓” =3V,保 ,56之交點’成為資,料「/5路之電壓減少特性2與特 :貧料「〇」胃入動作上之」:持動作點3。另—方面, 3. 5V,保持動作中之Vf掸 電體之施加電壓Vf = 特性1與特性56之交點σϋ,電滯迴路之電堡增加 由圖1 3,因為資料「「貝π 「0」保持動作點4。 =-0.8V,則Vf = —〇 8ν,^保持期間中之動作點為Vx 為資料「〇」保持期間中之〇V,則Vi = 〇.8V,因 〇.7v、由vg = ov,則Vi = 點為Vx = 〇,n,則Vf = 狀態,由於Vi &gt; vth,FET ·7ν。於/料「1」保持 態,由於n〈Vth,FET則於貧料「〇」保持狀 〇 5V最之後電上制±電極—為汗置狀態,於汲極電極上施加Vd = 限電壓值(Vi &gt; vth/|ljmH s 高於阳之臨 ET成為導通,汲極電流合古 。另一方面,若漂移電極之電壓低於FET之臨限電曰壓”值 1 &lt; Vth),則FET成為關閉,汲極電流則不會流動。 於此習知例,由於以資料r丨」保持狀態之動作點 解析所求得之Vi = (K8V遠高於Vth = 0.5V,FeFET便能讀 出貢料「1」。同時,由於以資料「0」保持狀態之動貝Vx = -3.5V, so Vf = -3 5V, ± -H.5V holds. From the formula ⑴, when Vi = state, this Vx saves the data in this state, although the control electrode is floating, the voltage drop caused by the leakage of the ferroelectric and FET, and the voltage of each electrode is light. It can be tolerated that the control electrode voltage is almost 0V. Because the control electrode is in a floating state, it is assumed that Vg = ov (vi = Vg _ and -Vx). In Fig. 1, 3 shows the "pair" characteristic of the operating state in the data holding state, 2 and Qi pair (Vg-Vf) characteristic 56. After writing the above-mentioned binary data, the operating point is determined in accordance with the electric hysteresis characteristic of the ferroelectric. That is, by MM, Q〇l 05477 V. Description of the invention (5) Data "1, said correction" The clock in the writing action + 2 Vf in the operation is reduced, the voltage applied to the electricity, belt: mouth body " = 3V, guaranty, the intersection of 56 'becomes the capital, and it is expected that the voltage reduction characteristic of the / 5 channel 2 and the special feature: lean material "0" on the stomach movement action ": hold action point 3. On the other hand, at 3.5V, the voltage applied to the Vf in the holding operation is Vf = the intersection point σ 特性 of characteristic 1 and characteristic 56. The electric fort of the hysteresis loop is increased by Figure 13 because the data is "" 贝 π "0 "Keep action point 4. = -0.8V, then Vf = -〇8ν, ^ The operating point in the holding period is Vx is 〇V in the data "0" holding period, then Vi = 0.8V, because 0.7v, from vg = ov, Then Vi = point is Vx = 〇, n, then Vf = state, because Vi &gt; vth, FET · 7ν. The state of “1” is maintained, because n <Vth, the FET remains in the state of “lean” 0. The last 5V is electrically turned on and the electrode is in a sweat state, and Vd is applied to the drain electrode. (Vi &gt; vth / | ljmH s becomes higher than Yang Zhilin ET to be turned on, and the drain current is combined. On the other hand, if the voltage of the drift electrode is lower than the threshold voltage of the FET "value 1 &lt; Vth), The FET will be turned off, and the drain current will not flow. In this conventional example, because Vi = (K8V is much higher than Vth = 0.5V, which is obtained by analyzing the operating point with the data r 丨 "held in the state, the FeFET can Read the tribute "1". At the same time, because of the state of keeping the data "0"

作點解析所求得之Vi = —0.7V遠低於Vth = 0.5V,FeFET 便能讀出資料 「0」。 於T. Nakamura 等人,&quot;ElectricalWhen the point analysis obtained Vi = —0.7V is far lower than Vth = 0.5V, the FeFET can read the data “0”. In T. Nakamura et al., &Quot; Electrical

Characteristics of MFMIS FET Using STN Films'1,Characteristics of MFMIS FET Using STN Films'1,

第10頁 五、發明說明(6) r:;::r;r\:;/i^r9r;57vsDM98·208' 。先前所述之動作模擬係假設Vg g 0:為例,之 此種情況下可以得到相同的結果。 之故於 _IΪ,對於利用鐵電體之不揮發性記憶元件,進# # =作點改變,面臨了無法正確地將資料讀出了=: =其亦即,雖詩f料保持動作求得” , ί通時之條件Vi〉vth之變動容許值,亦即,為 電:電Γ?ν進f貧Γ1」之寫入、讀出動作,漂移閉極 =:(.之變動容許值只有&quot;V。因此,由於降低 件一 …W3V,便無法滿順於導通 將面臨無法正確地將資料讀出之問題。 νΐ·&quot; = '°·7ν,^ n ni Vl〈 VU之變動容許值為12¥。 &quot; :呆持狀態下,由於FET於資料「1」之情況 =複數個FeFET之及極電極連接於位元線之記 5%Hl °卢二久如已公開於特開平5-205487號與特開平 :^ i (二者均是1 993年8月13曰發 :個/^Ϊίϊ電晶體附加於沒極電極上。於此,將複Page 10 V. Description of the invention (6) r:; :: r; r \:; / i ^ r9r; 57vsDM98 · 208 '. The action simulation previously described assumes Vg g 0: as an example, in which case the same result can be obtained. For _IΪ, for non-volatile memory elements using ferroelectrics, enter # # = make a point change and face the inability to read the data correctly =: = That is, although the poem f keeps the action required ””, The permissible variation of the condition Vi> vth at the time of ON, that is, the write and read operations of electricity: electricity Γ? Ν into f poor Γ1 ”, drift closed pole =: (. There is only &quot; V. Therefore, because the lower part ... W3V, you will not be able to follow the continuity and you will not be able to read the data correctly. Νΐ · &quot; = '° · 7ν, ^ n ni Vl <VU changes The allowable value is 12 ¥. &Quot;: In the state of holding, because the FET is in the data "1" = the number of FeFETs and the electrode connected to the bit line is 5% Hl ° Lu Erjiu has been disclosed in the special Kaiping No. 5-205487 and Special Kaiping: ^ i (both are issued on August 13, 1 993: a / ^ Ϊίϊ transistor is attached to the electrode electrode. Here, we will repeat

ί :二Λ 、陣形产之半導體記憶裝置,檢測FeFET 、二齡S於#二而進行讀出動作時,必須利用選擇電晶體 狀態之其他的FeFET没極電流之故,將面 S品無法進仃積體化之問題。 WU859ί: When a semiconductor memory device of two Λ and formation is used to detect the read operation of FeFET and second-generation S in # 2, it is necessary to use the other FeFET selected in the state of the transistor to have no pole current, so that the surface S product cannot be imported. The problem of integration. WU859

[發明概要] 為了解決上述之課題,本&amp; 體記憶褒置之保持特性dc提供-種半導 法。 憨的貝枓頊出之最佳電壓設定方 本發明係有關於半導體記憶 成對的電極與鐵電體所構成的電容,’由 :極極連接於上述場效應電晶體之間極 之閘極電極,將上述電容之另^為上述%效應電晶體 ,_ ^ ^ 肘上江電谷之另一側電極做為控制電極,夢 而改變變上述鐵電體之偏極化, ,^上^%效應電晶體之通道阻抗,由於上述通道阻抗 ^ π低而表示二值資料之半導體記憶裝置之資料讀出方 忐,其特徵為:若上述場效應電晶體之臨 正電壓於上述控制電極,若上述場效==臨 限電壓值為負,則施加負電壓於上述控制電極。根據該 料讀出方法,於讀出動作時,將特定之電壓(Vread)施、 加於控制電極,習知的資料讀出方法,對於關閉之變動容 許值為1.2V,而對於導通之變動容許值則較小,為〇 3V, 能消除導通與關閉之變動容許值的差值。亦即,為了滿足 導通時之條件Vi &gt; vth而能擴大變動容許值,進而能改善 保持特性。 b 口 本發明係有關於上述資料讀出方法,若上述場效應電 晶體之臨限電壓值為正,上述電容之殘留偏極化值於〇庫[Summary of the Invention] In order to solve the above-mentioned problems, the present invention provides a semiconducting method for retaining characteristics dc of a body memory device. The present invention relates to a capacitor composed of a pair of electrodes of a semiconductor memory and a ferroelectric body. The electrode is connected to a gate electrode between the above-mentioned field effect transistors. Let the other capacitor of the above capacitor be the above-mentioned% effect transistor, _ ^ ^ The other electrode on the Elbow River Valley is used as the control electrode, and change the polarization of the ferroelectric body, ^^^ The channel impedance of the effect transistor, because the above-mentioned channel impedance ^ π is low, represents the data readout of the semiconductor memory device of binary data. It is characterized in that if the positive voltage of the field-effect transistor is on the control electrode, if If the above field effect == threshold voltage is negative, a negative voltage is applied to the control electrode. According to the material readout method, a specific voltage (Vread) is applied to the control electrode during the readout operation. The conventional data readout method has a permissible value of 1.2V for the change of turn-off and a change of the turn-on. The allowable value is small, which is 0 3V, which can eliminate the difference between the allowable value of the on and off variations. That is, in order to satisfy the condition Vi &gt; vth at the time of turn-on, the allowable value of variation can be increased, and the retention characteristics can be improved. b. The present invention relates to the above-mentioned data reading method. If the threshold voltage value of the field-effect transistor is positive, the residual polarization value of the capacitor is stored in the library.

修正 曰 號 9010M77 五、發明說明(8) 平方公分之狀態7 ’則將電極間 文Π之臨,施加於上述場效應電晶體之間 值之雷Η e f生相虽於對通道所感應產生之電荷偏極化 加,施加於上述㈣電極應電晶體之臨限電壓值相 KG為”電壓I;若上述場效應電晶體之臨 : = 電容之殘留偏極化值於0,倫/平方 值改變時之臨限電壓方“ „開發減少,使偏極化 :門ίϋ ί於對通道所感應產生之電荷偏極化值:電 = 述場效應電晶體之臨限電壓值相加,施 心±20%以内之電壓值。最好疋所得到之電隸為中 ㈣m敘Λ,削1s MFeFET之鐵電體電容之偏極化值 二! 狀態下,利用電極間電壓⑴)由 0V起&amp;加時之Pf對”特性與FETiQi對^ 广亦即降低偏 之最差炀況假汉為Pf = 〇庫倫/平方公 ^ ^ 讀出電壓。由此,於資料U」保持動刀作之中W吏而^疋 化值降低,FET必須於導通時才能達到Pf = 能正確地將資料讀出。還有,考量 便 準確性,施加於控制電極之讀出電壓™:^及偏極化之不 光罩之疊對偏移、㈣之不準 =度之不準碟性、 離子植入量之不準確 第13頁 曰 一修正 差號 90105477_月 五、發明說明(9) 性、受損等之影響,而加上土 15% ^ r〇/ 的不準確性而設定鐵電體之偏極化值不切性,由於±5% 法,::對月係有關於半導體記憶裳置之資料寫入方 體所組: 鐵'體:構成的電容,與場效應電晶 之閘極ΐ極ΐ?谷:電極連接於上述場效應電晶體 電晶體之閑極電極,將上述電容之另录作為上述场效應 極化,: ::? 述控制電極以改變上述鐵電體之偏 道阻抗場效應電晶體之通道阻抗,由於上述通 入方示:值資料之半導體記憶裝置之資料寫 加電壓後述控制電極施加之電廢值,於施 於非導通狀能、Π錄態’ 1述場效應電晶胃成為處 C之範圍内。根據該資料寫入方法,於資料 電流並”』動之各保持狀態,FET成為關閉狀態,汲極 為藉憶裝置之驅動方法,其特徵 行資料ΐιΓί 導體記憶裝置之資料寫入方法而進 根據★亥驅動方ΐ,施加重新設定的電壓於上述控制電極。 動方法’☆資料寫入動作之後,會立刻向保持動 ^本i ί於依本發明的上述資料寫入方法而進行資料寫入 記憶裝置,進行其資料讀出之時,上述 控制電極,並且於上、+、ia I ΐ I最好施加正電壓於上述 、上述%效應電晶體施加低於臨限電壓值Correction code: 9010M77 V. Description of the invention (8) The state of 7 cm² will bring the interfacial electrode Π to the thunder ef which is the value between the above-mentioned field-effect transistors. The charge polarization is added, and the threshold voltage phase KG applied to the above-mentioned 应 electrode should be “voltage I; if the above-mentioned field-effect transistor is near: = the residual polarization value of the capacitor is 0, and the value is the square / square value. Threshold voltage at the time of change "„ Development is reduced to make the polarization: gate ίϋ ί the value of the polarization of the charge induced by the channel: electricity = sum the threshold voltage values of the field effect transistor, add heart The voltage value within ± 20%. It is best to get the polarization value of ㈣m 叙 Λ, cut the 1s MFeFET ferroelectric capacitor's polarization value 2! In the state, use the voltage between the electrodes ⑴) from 0V & amp The characteristic of "overtime Pf pair" and the FETiQi pair are the widest, which is the worst case for reducing bias. The dummy is Pf = 0 Coulomb / square meter ^ ^ read voltage. As a result, while the data U ”keeps moving, the value is reduced, and the FET must be turned on to reach Pf = the data can be read correctly. Also, consider the accuracy of the readout voltage ™ applied to the control electrode: ^ and the bias of the polarized non-masked stack offset, the inaccuracy of the degree = the inaccuracy of the degree, the inaccuracy of the ion implantation Accurate on page 13 is a correction of the difference 90105477_Fifth, the description of the invention (9) the impact of sex, damage, etc., plus the 15% ^ r〇 / inaccuracy to set the polarization of the ferroelectric The value is not tangible, due to the ± 5% method :: For the month, there is information about the semiconductor memory device written into the cube. Group: Iron 'body: the capacitor formed, and the gate electrode of the field effect transistor. Valley: The electrode is connected to the free-electrode of the above field-effect transistor, and the above-mentioned capacitor is recorded as the above-mentioned field-effect polarization: ::? The control electrode changes the channel impedance of the above-mentioned ferroelectric's off-channel impedance field-effect transistor. Because the above access method shows: the value of the semiconductor memory device data is written and the voltage is applied. In the non-conducting state, the state of the field effect transistor is within the range of C. According to the data writing method, the FET is turned off in each of the holding states where the data current is turned on, and the driving method of the memory device is drawn. Its characteristic data is based on the data writing method of the conductor memory device. The driver drives Fang to apply a reset voltage to the above control electrode. Operation method '☆ After the data writing operation, it will be moved to the holding position immediately. I 本 i According to the above-mentioned data writing method of the present invention, data writing and memory are performed. When the device reads its data, the control electrode above, +, and ia I ΐ I should preferably apply a positive voltage to the above, the% effect transistor above the threshold voltage value

第14頁 月 曰 修正 案藏 90105477 五、發明說明(10) 的電壓;而上述場效應 下,施加負電麼於上述控制電極為負的,月況 f施加高於臨限電遷值的電壓。根據該資電; _,於讀出時,若資料保f」狀:導=關= 「〇」為關閉動作,則能正確地將資料K作…貝枓 鐵電可以適用於由上述成對的電極與 曰心㈣之PW *、電極連接於控制電極,上述第2場效應電 :狳夕:二1極連接於第1字元線,汲極電極連接於第2字 極雷炻广丫列。因此’於汲極電極/位元線間(或是源 次線間)即使省略選擇電晶體也可以正確地將 貝枓頃出,因而能提昇積集之程度。 # μ,據t上所述之本發明,不僅能擴大FeFET之讀出動 、電壓k動容許值,也可改變鐵電體電容之偏極化 八太即使在最差情況下,直到偏極化成為◦庫倫/平方公 二”、、止,仍能正確地進行讀出動作。同時,也可以省略汲 極電極與位元線間之選擇電晶體。 [實施例之詳細說明] 以^下針對於本發明之實施例,參照著圖面加以說明。 C貫施例1) 第15頁 490859Page 14 Month Amendment Collection 90105477 V. The voltage of the invention (10); and under the above-mentioned field effect, the application of negative electricity is negative to the above control electrode, and the monthly condition f applies a voltage higher than the threshold electrical migration value. According to the information; _, when reading, if the data is kept f "state: guide = off =" 〇 "is the closing action, the data K can be correctly made ... The electrode is connected to the PW * of the heart, and the electrode is connected to the control electrode. The above-mentioned second field-effect voltage is: the first and second electrodes are connected to the first word line, and the drain electrode is connected to the second word. Column. Therefore, even if the selection of the transistor is omitted between the drain electrode / bit line (or between the source line and the sub-line), the shell can be accurately extracted, and the degree of accumulation can be improved. # μ, according to the invention described above, not only can the allowable values of the read and voltage k movements of the FeFET be changed, but also the polarization polarization of the ferroelectric capacitor can be changed. Even in the worst case, the polarization polarization Become Coulomb / square square ", and can still read correctly. At the same time, the selection transistor between the drain electrode and the bit line can be omitted. [Detailed description of the embodiment] The embodiment of the present invention will be described with reference to the drawings. C. Embodiment 1) Page 15 490859

ι i f之貝鈀例1的半導體記憶裝置,利用圖9以說明 :上與習知技術之構造是相同的,其等效電路圖 ,與圖1M目同。同時’寫人動作、保持動作也與習知技術 相同。 本發明之實施例1的以心了與f知的FeFET之間的不同 點為讀出動作時之驅動電壓Vread的設定方法。亦即,若 場效應電晶體之臨限電壓值vth為正,則藉由施加正電壓 =控制電極,若場效應電晶體之臨限電壓值vth為負,則 藉由施加負電壓於控制電極,設定驅動電壓值Vread。 於此’若場效應電晶體之臨限電壓值Vth為正,則施 加Vg = 〇. 5V的電壓於控制電極而進行讀出動作。此時, 係於資料「1」讀出動作,I ds因而增加。 由於與習知例具有同樣之FeFET構造,針對以下列之 規格所製作的FeFET,鐵電體電容之電極面積為2.8/zm X 2 · 8 // m ’ F E T之閘極長為〇 · 8 // m,閘極寬為4 # m,以I d s做 為指標而進行性能評估。 圖1係顯示資料 「1」與資料「0」之I d s比(導通 /關閉比)與Vg之關係。習知的驅動方法(Vg = 〇V)The semiconductor memory device of Example 1 of palladium and palladium will be described with reference to FIG. 9: the structure of the conventional and conventional technologies is the same, and its equivalent circuit diagram is the same as that of FIG. 1M. At the same time, the writing and holding actions are also the same as the conventional techniques. In the first embodiment of the present invention, the difference between the FeFET and the known FeFET is the method of setting the drive voltage Vread during the read operation. That is, if the threshold voltage value vth of the field effect transistor is positive, then by applying a positive voltage = control electrode, if the threshold voltage value vth of the field effect transistor is negative, then by applying a negative voltage to the control electrode , Set the drive voltage value Vread. Here, if the threshold voltage value Vth of the field effect transistor is positive, a voltage of Vg = 0.5 V is applied to the control electrode to perform a read operation. At this time, due to the reading operation of the data "1", I ds increases accordingly. Because it has the same FeFET structure as the conventional example, the electrode area of the ferroelectric capacitor is 2.8 / zm X 2 · 8 // for the FeFET fabricated with the following specifications. The gate length of the m FET is 0 · 8 / / m, the gate width is 4 # m, and I ds is used as an index for performance evaluation. Figure 1 shows the relationship between the I d s ratio (on / off ratio) and Vg of the data "1" and the data "0". Conventional driving method (Vg = 〇V)

時,如圖1所示之交點a,I ds比為1 〇4。對於此,根據本實 施例1 (Vg = 〇· 5V),如圖1所示之交點b,Ids比為5 X 1 〇5 °亦即,於控制電極施加〇 · 5 V之電壓時,初期之I ds比 為5 0倍。 將該樣品於室溫下長時間放置而調查I ds之變化。其 結果為,放置1 〇6秒後,資料 「1」 之I ds會降低至約其At this time, the intersection point a and the I ds ratio are 104 as shown in FIG. 1. Regarding this, according to Example 1 (Vg = 0.5 V), as shown in the intersection point b shown in FIG. 1, the Ids ratio is 5 X 1 0.5 °, that is, when a voltage of 0.5 V is applied to the control electrode, the initial stage The I ds ratio is 50 times. This sample was left at room temperature for a long time to investigate the change in I ds. As a result, the I ds of the data “1” will be reduced to about the same after being left for 106 seconds.

第16頁 案號 90105477 五、發明說明(12) 值之1/10。但疋,由於初期之Ids較知 到動作異常為止之壽命也較習知驅動為長。 直 接著,與習知例同樣的手法以模擬本實施例 作,並驗證其妥當性。 貝出動 由=入動作、保持動作與習知技術相同,與圖 T,圖11B、圖12A ’圖12B及圖13所示者進行同樣的動 、兄為si;所於保持動作狀態之動作點,資# Γ1」之情 =4圖13所不之交點3,資料「〇」之情況為圖u所示之 •^屮^二&amp;例係使用圖2以說明有關於從保持狀態開始之 二=:2於二2:1是增加鐵電體電容之電壓而得到的Pf :.3 Λ ’ j疋「減少鐵電體電容之電壓而得到的P f對V f特 Ϊ動;V /,1」料之動作點;4是資料%保待 性6 =。…彻之㈣(Vg - Vf) ί 次料7 =點3開始’增加電壓時之Pd對Vf特性;7是 :。」頊出之動作點;8是資料「〇」讀出之動作 牛,H實施例1讀出動作時之驅動電驗ead的設定方 之正電壓而進行讀出動作為正,則猎由施加於控制電极 〇例=,如圖2所示,施加Vg = 〇. 5V下,讀出資料 厂 」二從保持狀態之Vg〜〇v開始,電壓增加。 的/對v?:動作點4係處於增加鐵電體電容之電壓所得到 的⑴川特性U之故,Pf,Page 16 Case No. 90105477 V. Description of the invention (12) 1/10 of the value. However, since the Ids at the initial stage are better known, the lifespan up to the abnormal operation is longer than the conventional drive. Next, the same method as in the conventional example was used to simulate the operation of this embodiment, and its validity was verified. The start-up movement and the holding movement are the same as the conventional technique, and the same movement is performed as shown in Figure T, Figure 11B, Figure 12A, Figure 12B, and Figure 13;情 # Γ1 ”sentiment = 4 intersection point 3 shown in FIG. 13 and the case of data“ 〇 ”is shown in figure u. ^ 二 ^ 二 &amp; For example, use FIG. 2 to explain about Two =: 2 and two 2: 1 are Pf obtained by increasing the voltage of the ferroelectric capacitor: .3 Λ 'j 疋 "P f obtained by reducing the voltage of the ferroelectric capacitor is particularly responsive to V f; V / , "1" is the expected operating point; 4 is the data% reserve 6 =. … Tochi no Kou (Vg-Vf) ί times material 7 = start at point 3 ′ Pd vs. Vf characteristic when increasing voltage; 7 is:. The action points that were identified; 8 are the action cows that read out the data "0", H. The positive voltage of the setting side of the drive test ead during the read operation of Example 1 is positive, and the read action is positive. For example, as shown in FIG. 2, when the control electrode is applied, as shown in FIG. 2, under the application of Vg = 0.5V, the data factory is read out. The voltage is increased from Vg ~ 0v in the holding state. / Pair v ?: The operating point 4 is at the reason of the characteristics of Ugawa, U, obtained by increasing the voltage of the ferroelectric capacitor, Pf,

五、發明說明(13)V. Description of Invention (13)

Vf)特性5之父點即為資料「〇」讀出動作點8。由圖2, 由於動作點8可讀出vx = 〇·8ν之故,vf = — vx = 〜〇.8V , Vi= Vg 〜Vx = —〇 3V 。 接著’於貧料「j」讀出動作時,資料r丨」保持 動作點3係處於減少鐵電體電容之電壓而所得到之p f對v f 特性2上之故’從動作點3開始,增加電壓時,不得不求出 鐵電體電容之Pf對Vf特性。 如此’從電滯迴路之低於飽和偏極化值(Ps)的狀態 開始’改變電壓時之鐵電體的舉動,於s· L. Mi丨ler等 人,Mode 1ing ferroe 1 ectric capacitor switching with asymmetric nonperiodic input signals and arbitrary initial conditions”, Journal 〇f Appiied Physics, 70 (5), pp· 2849-2860, Sep· 1991 中已闡 明。利用M l 1 1 e r等人之手法,將飽和電滞迴路之偏極化 值設為Psat (Psat事實上與直至目前為止所討論之所用以 表示的偏極化值P f相同),處於飽和電滯迴路之内側之偏 極化值設為P d (偏極化值處於飽和電滞迴路之内側時,為 了與Psat區別而稱為Pd)時,以式(4)表示以常數r對 於鐵電體所施加的電場E之偏微分相結合。 d ?d d E = Γ · d Psat / d E (4) 於此,以式(5)求得Γ。但是,若電壓增加時,則 f = +1 ;若電壓減少時,則f -1。 Γ =1-tanh[ {(Pd-Psat)/( f · Ps-Pd) }i/2 ] (5) 利用此M i 11 e r等人之手法,求出從動作點3開始,增Vf) The father point of characteristic 5 is the data "0" read operation point 8. From FIG. 2, since the operating point 8 can read vx = 〇 · 8ν, vf = — vx = ~ 〇.8V, Vi = Vg ~ Vx = — 〇 3V. Next, "When the lean material" j "reads the action, the data r 丨" keeps the operating point 3 at the pf vs. vf characteristic 2 obtained by reducing the voltage of the ferroelectric capacitance. "Starting from the operating point 3, increasing For voltage, the Pf vs. Vf characteristic of the ferroelectric capacitor has to be determined. In this way, the behavior of the ferroelectric body when changing the voltage "starting from the state of the hysteresis loop below the saturation polarization value (Ps)" is described in S.L. Mierer et al., Mode 1ing ferroe 1 ectric capacitor switching with asymmetric nonperiodic input signals and arbitrary initial conditions ", as explained in Journal 〇f Appiied Physics, 70 (5), pp · 2849-2860, Sep · 1991. Using the method of M l 1 1 er et al. The polarization value is set to Psat (Psat is actually the same as the polarization value P f that has been discussed so far), and the polarization value inside the saturated hysteresis loop is set to P d (bias When the polarization value is inside the saturated hysteresis loop, it is called Pd for the purpose of distinguishing it from Psat, and the partial differential of the constant r to the electric field E applied to the ferroelectric is expressed by Equation (4). D? Dd E = Γ · d Psat / d E (4) Here, Γ can be obtained by equation (5). However, if the voltage increases, f = +1; if the voltage decreases, f -1. Γ = 1 -tanh [{(Pd-Psat) / (f · Ps-Pd)} i / 2] (5) Using this method of M i 11 er and others, find out from 3 point for start, increase

第18頁Page 18

490859 修正490859 fix

案號 90105477 五、發明說明(14) 性與 加電壓時之Pd對Vf特性,為圖2之 FET之Qi對(Vg - Vf)特性5 6 3亥Pd對”特 τ 1王3之乂點為資料 「1 ^ 作Τ。由圖2,於動作點7可以讀出Vx : -0 4VU買出動 …Vx = 0.4V、Vi = Vg _ νχ = uv。.之故,Vf 由於FET為導通之條件㈣藉n m, 1」讀出之變動容許值成為Vi _ _ 〇 9、、= 0.4V。亦即’與習知的讀出方法之資料ri. :5V = 容許值G.3V相比較,變動容許值僅能擴大o.1V。同時k,動由 於滿足FET為關閉之條件係資料「〇」讀出之變^ : 成為Vth _ Vi = 0· 5V - (―〇· 3ν) = 〇· 8V。 卉值 如此,雖然習知導通之變動容許值(〇· 3V)遠小於 閉之變動容許值U.2V),藉由施加Vg = 〇·5 V而讀出、/ 能擴大導通之變動容許值,而能提供完全適用於保持特性 之FeFET 〇 還有’本發明之實施例1係利用Vth = 〇· 5V iN通道型 FET,施加於控制電極Vg = 〇· 5V之電壓而讀出。例如, Vth &lt; 0V之空乏型的N通道型FET時,遵照直至目前為止之 所論述,關閉之變動容許值比導通之變動容許值為小是可 以理解的。如此情況下,藉由施加於控制電極之負電壓 (Vg &lt; ΟV),能擴大關閉之變動容許值。同時,對於p通道 型FET時也相同。 (實施例2 ) 本發明之實施例2的半導體記憶裝置,與習知例及實Case No. 90105477 V. Description of the invention (14) The characteristics of Pd vs. Vf during application of voltage and voltage are the characteristics of Qi pair (Vg-Vf) of the FET in Figure 2 For the data "1 ^ as T. From Figure 2, Vx can be read out at action point 7: 4VU buy out ... Vx = 0.4V, Vi = Vg _ νχ = uv. Therefore, Vf because the FET is turned on The allowable value of the readout under the condition ㈣nm, 1 ″ becomes Vi _ _ 09, = 0.4V. That is, compared with the data of the conventional reading method ri.: 5V = allowable value G.3V, the allowable value of variation can only be increased by o.1V. At the same time, the change in the reading of the data "〇" due to the condition that the FET is turned off ^: becomes Vth_Vi = 0.5V-(-〇 · 3ν) = 0.8V. The value is so. Although the permissible variation of the conduction (0.3 V) is much smaller than the permissible variation of the closed U. 2 V), it can be read by applying Vg = 0.5 V, and the permissible variation of the conduction can be enlarged. In addition, a FeFET that is completely suitable for holding characteristics can be provided. Also, the first embodiment of the present invention uses a Vth = 0.5V iN channel FET, which is read by applying a voltage of the control electrode Vg = 0.5V. For example, in the case of a Vth &lt; 0V empty N-channel FET, it is understandable that the permissible variation value of the turn-off is smaller than the permissible variation value of the on-state according to the discussion so far. In this case, the allowable value of the variation of the shutdown can be increased by the negative voltage (Vg &lt; OV) applied to the control electrode. The same applies to p-channel FETs. (Embodiment 2) A semiconductor memory device according to Embodiment 2 of the present invention, and a conventional example and implementation

第19頁 490859 ----案號 90105477 五、發明說明(15) 年月曰_Page 19 490859 ---- Case No. 90105477 V. Description of Invention (15)

施例1之構造與等效電路圖相同。同時,寫入 動作也與習知例及實施例1相同。 一保持 ^ 本發明之實施例2與實施例1之不同點為··於每 讀出動作時之驅動電MVread之設定方法係FET之二:例1 ,Vth為正的情況下,藉由施加正電壓於控制電極而^電壓 =出動作,實施例2的特徵係更進一步地於讀出動护仃 施=於控制電極之電壓值Vread之最佳設定方法。亦/ 所施加於控制電極之電壓值Vread,若場效應電晶# /故 壓值Vth為正’上述電容之殘留偏極化值為〇庫偷/平 方公分的狀態下,則將電極間電壓由〇v起增加,而使时偏極 化值改變時之臨限電壓vti施加於場效應電晶體之閘極電 極時,使產生相當於對通道所感應產生之電荷偏極化值之 電極間電壓值,與上述場效應電晶體之臨限電壓值相加, 而得到之電壓值為中心± 20%以内之電壓值。同時,所施 力^於控制電極之電壓值Vread,若場效應電晶體之臨限電 =值Vth為負,上述電容之殘留偏極化值為〇庫倫/平方公 刀的狀悲下,則將電極間電壓由〇 V起減少,而使偏極化值 改變日守之臨限電壓v th施加於上述場效應電晶體之閘極電 極時,使產生相當於對通道所感應產生之電荷偏極化值之 電極間電壓值,與上述場效應電晶體之臨限電壓值相加, 而得到之電壓值為中心土 2 0 %以内之電壓值。 使用圖3A〜C以說明本發明實施例2之讀出動作之電壓 設定方法。 圖3A之21係鐵電體電容之偏極化值為〇庫倫/平方公The structure of the first embodiment is the same as that of the equivalent circuit diagram. At the same time, the writing operation is the same as in the conventional example and the first embodiment. One hold ^ The difference between the second embodiment of the present invention and the first embodiment is that the setting method of the drive voltage MVread at each read operation is the second of the FET: Example 1, when Vth is positive, by applying The positive voltage is at the control electrode and ^ voltage = output action. The characteristic of the embodiment 2 is further the best setting method for reading the dynamic protection application = the voltage value Vread at the control electrode. Also, the voltage value Vread applied to the control electrode, if the field-effect transistor # / thus the voltage value Vth is positive, the residual polarization value of the capacitor is 0, and the voltage between the electrodes is changed. Increased from 0v, so that the threshold voltage vti when the time polarization value is changed is applied to the gate electrode of the field effect transistor, so that between the electrodes that generate the polarization value corresponding to the charge induced by the channel The voltage value is added to the threshold voltage value of the above-mentioned field effect transistor, and the obtained voltage value is a voltage value within the center ± 20%. At the same time, the applied force is the voltage value Vread of the control electrode. If the threshold voltage of the field effect transistor = the value Vth is negative, the residual polarization value of the capacitor described above will be 0 coulomb / square knife. When the inter-electrode voltage is reduced from 0V, and the polarization value is changed, the threshold voltage vth of the day guard is applied to the gate electrode of the above-mentioned field-effect transistor, so that a charge equivalent to that induced by the channel is generated. The voltage value between the electrodes and the threshold voltage value of the field effect transistor are added together, and the obtained voltage value is a voltage value within 20% of the center soil. 3A to 3C are used to explain the voltage setting method of the read operation according to the second embodiment of the present invention. The polarization value of the 21 series ferroelectric capacitor in Figure 3A is 0 coulomb / square centimeter.

钎观59 ---9010547J--^^月 日 條正_ 五、發明說明(16) ' -- 分的狀態開始,增加時之Pd _Vf特性。於FeFET之鐵電體 電容區之偏極化值,在保持動作中降低時,最差的情況是 偏極化到達〇庫倫/平方公分狀態的情形。圖^之21係從〇 C/cm2的狀態開始,增加電壓時之以對”特性,由此可求 出於鐵電體電容區之Pd對Vf特性值。 圖3B之22係FET之Qi對Vi之特性。於此特性22,於臨 限電壓值(Vth = 0.5V)之電荷設定為Qti (= 〇 2庫倫 /平方公分)。於鐵電體電容,由圖3人之21求出而可以得到 與此Qti相等之偏極化值之電壓值,該電壓值設定為vtf (=0·05V)。Observation 59 --- 9010547J-^^ month day Article _ V. Description of the invention (16) '-The state of the minute starts, and the Pd_Vf characteristic when it increases. When the polarization value of the ferroelectric capacitance region of the FeFET is reduced during the holding operation, the worst case is when the polarization reaches a state of 0 Coulomb / cm 2. Figure 21 shows the characteristics of Pd vs. Vf from the ferroelectric capacitor region, starting from the state of 0C / cm2. The characteristic of Pd vs. Vf can be obtained from the ferroelectric capacitor region. Figure 3B. The characteristic of Vi. In this characteristic 22, the charge at the threshold voltage value (Vth = 0.5V) is set to Qti (= 〇2 Coulomb / cm 2). For the ferroelectric capacitor, it is obtained from 21 of Figure 3 The voltage value of the polarization value equal to this Qti can be obtained, and the voltage value is set to vtf (= 0.05V).

本貫施例2,以Vread = Vtf + Vth = 0.05V + G.5V =〇· 55V所得到的電壓值施加於控制電極而進行讀出動 作。此Vtf、Vth之值,考量臨限電壓及偏極化的不準確 性,由於Vth之不準確性為± 15%,Vread之不準確性為 ±20% 左右。此時,〇.〇475v〈 Vtf &lt; 〇〇525v,〇425v &lt;In the second embodiment, a voltage value obtained by Vread = Vtf + Vth = 0.05V + G. 5V = 0.55 V is applied to the control electrode to perform a read operation. The values of Vtf and Vth take into account the inaccuracy of the threshold voltage and the polarization. Because the inaccuracy of Vth is ± 15%, the inaccuracy of Vread is about ± 20%. At this time, 0.00475v <Vtf &lt; 〇5255v, 〇425v &lt;

Vth &lt; 0. 575V 成立。 於圖3C之曲線23係顯示以vg = Vread做為施加此讀出 電C日守之FET的Qi對(Vg - vf )特性。亦即,圖3C係表示 最,的情況是從偏極化〇庫倫/平方公分之狀態開始,進 行讀出\動\作時之^作狀態。根據此方法,從偏極化Q庫倫 /平方A刀之狀悲開始,於增加電壓時之μ對η特性2 1與 FET的Qi對(Vg - Vf)特性23之交點,施加於FET之漂移閘 極電,之電壓Vi為臨限電壓值vth (Vi = vth)。於此,假 設該最差的情況時之動作點為^ = Vx〇 (= vtf)。Vth &lt; 0. 575V was established. Curve 23 in Fig. 3C shows the characteristics of the Qi pair (Vg-vf) of the FET with vg = Vread applied to this readout. That is to say, Fig. 3C shows the case where the state of the operation at the time of readout \ operation \ operation is started from the state of polarization of 0 Coulomb / cm 2. According to this method, starting from the state of polarized Q Coulomb / square A knife, the point where the μ pair η characteristic 21 and the Qi pair (Vg-Vf) characteristic 23 of the FET at the time of increasing the voltage are applied to the FET drift Gate voltage, the voltage Vi is the threshold voltage value vth (Vi = vth). Here, it is assumed that the operating point in the worst case is ^ = Vx〇 (= vtf).

490859 __案號90105477_年月曰__ 五、發明說明(17) 若遠大於偏極化之最差情況(Pd &gt; 0庫倫/平方公 分)’則進行讀出動作時之Pd對Vf特性處於較圖3C之21更 上側之故,與FET之Qi對(Vg - Vf)特性23之交點,處於 車父圖3C之2 1更左側。若該交點之電壓假設為yx 1,則1 &lt;490859 __Case No. 90105477_ 年月 月 __ V. Description of the invention (17) If it is far greater than the worst case of polarization (Pd &gt; 0 Coulomb / cm²), then Pd vs. Vf when reading Because the characteristic is on the upper side than 21 in FIG. 3C, the intersection point with the characteristic 23 of the Qi pair (Vg-Vf) of the FET is on the left of the parent figure 2C in FIG. 3C. If the voltage at the intersection is assumed to be yx 1, then 1 &lt;

VxO 。因此,Vi 二 Vread - Vxl = Vtf + Vth - Vxl =VxO. Therefore, Vi 2 Vread-Vxl = Vtf + Vth-Vxl =

VxO + Vth - VX1 &gt; Vti成立,滿足FET為導通時之條件Vi &gt; V t h 〇 若遠小於偏極化之最差情況(Pd〈 〇庫倫/平方公 分)’則進行讀出動作時之Pd對Vf特性處於較圖3C之21更 下側之故,與FET之Qi對(Vg - Vf)特性23之交點,處於 較圖3C之21更右側。若該交點之電壓假設為Vx2,則^2 &gt;VxO + Vth-VX1 &gt; Vti is established, satisfying the conditions when the FET is on Vi &gt; V th 〇 If it is far less than the worst case of polarization (Pd <〇 Coulomb / cm 2) ' Because the Pd vs. Vf characteristic is lower than 21 of FIG. 3C, the intersection with the Qi pair (Vg-Vf) characteristic 23 of the FET is further to the right than 21 of FIG. 3C. If the voltage at the intersection is assumed to be Vx2, then ^ 2 &gt;

VxO 。因此,Vi = Vread — Vx2 = Vtf + Vth ~ =VxO. Therefore, Vi = Vread — Vx2 = Vtf + Vth ~ =

VxO + Vth - Vx2 &lt; Vth成立,滿足FET為關閉時之條件Vi &lt; V t i ° 如此一來,利用本實施例2的手法而讀出動作時,若 設定控制電極電壓為Vread,且若偏極化值即使改變也大 於最差之情況(Pd &gt; 〇 C/cm2),則必須滿足FET為導通時 之條件Vi &gt; vth,才能將資料「1」正確地讀出。同樣 地,若偏極化值即使改變也小於最差之情況(pd &lt; 〇庫 倫/平方公分),則必須滿足FET為關閉時之條件Vi &lt;VxO + Vth-Vx2 &lt; Vth is established and Vi &lt; V ti ° is satisfied when the FET is turned off. As a result, when the reading operation is performed by the method of the second embodiment, if the control electrode voltage is set to Vread, and if Even if the polarization value is changed more than the worst case (Pd &gt; 〇C / cm2), the condition "Vi" &gt; vth when the FET is turned on must be satisfied to read the data "1" correctly. Similarly, if the polarization value is smaller than the worst case even if it changes (pd &lt; 〇 Coulomb / cm 2), the condition Vi &lt;

Vth ’才能將資料「〇」正確地讀出。 還有’利用本實施例2以說明Vth = 〇·5ν之n通道型 FET。例如,vth&lt; 〇ViP通道型FET的情形,利用圖“所示 之由偏極化〇 C/cm2之狀態開始,減少電壓時的pd對”特Vth 'can read the data "〇" correctly. In addition, the second embodiment is used to explain an n-channel FET with Vth = 0.5V. For example, in the case of the vth &lt; 〇ViP channel FET, the figure "from the state of polarization polarization 0 C / cm2, pd pair when voltage is reduced" is used.

第22頁 490859 修正 案號90〗沾477 五、發明說明(18) 性24,圖巧所示之FET的Q^Vi特性25,與上述之方法相 同而能求得最佳的Vread。空乏型的N通道型FET的情形也 是相同的。 (實施例3 ) 於1|知的寫入動作’藉由施加於控制電極Vg = 土157 ^右的大電壓,即使在無施加電壓的狀態下,鐵電體為被 扁極化狀態,於源極/汲極間會有電流流動。 於本貫施例3寫入動作之控制電極,雖然施加於該控 ^,極之電壓值Vg,於施加電壓後之無施加電壓狀態,上 %效應電晶體處於成為處於非導通狀態之電壓值範圍 2。根據該組成,藉由施加非導通狀態之電壓值所在範圍 牵、之「小的寫入電壓於場效應電晶體之閘極電極電位v丨,資 二「/」及資料「0」之各保持狀態下,FET成為關、 才,汲極電流並不流動,同時於讀出時,藉由施加小 入電壓,而能正確地將資料讀出。 馬 本發明實施例3之半導體記憶裝置,與用以說明習 術的圖g之構造基本上是相同的,該等效電路圖也與 相同。另外,寫入動作、保持動作也與習知技術相同。 δ時,如貫施例1之說明,若場效應電晶體之臨限電壓值 th為正,則假設施加〇· 5V的電壓(vth)於控制 行讀出動作。 而進 ^雖然於習知的FeFET與實施例1,2之FeFET,寫入氣 係苑加Vg = ± 1 5V的電壓於控制電極,本實施例3則是寫作Page 22 490859 Amendment No. 90〗 Dian 477 V. Description of the Invention (18) Property 24, Q ^ Vi characteristic of the FET shown in the figure 25 is the same as the above method, and the best Vread can be obtained. The same applies to the empty N-channel FET. (Embodiment 3) In the writing operation known at 1 | ', with a large voltage applied to the control electrode Vg = 157 ^ ^, even in the state where no voltage is applied, the ferroelectric body is flat-polarized. Current flows between the source / drain. In the control electrode of the writing operation of Example 3, although the voltage value Vg of the control electrode is applied to the control electrode, there is no applied voltage state after the voltage is applied. The upper% effect transistor is in a non-conducting voltage value. Range 2. According to this composition, the "small write voltage is applied to the gate electrode potential v of the field-effect transistor by applying a voltage value in a non-conducting state, and each of the two" / "and the data" 0 "is maintained In the state, the FET is turned off and the drain current does not flow. At the same time, the data can be read correctly by applying a small input voltage during reading. The semiconductor memory device according to the third embodiment of the present invention is basically the same as the structure of figure g for explaining the practice, and the equivalent circuit diagram is also the same. The writing operation and the holding operation are also the same as in the conventional technique. At δ, as described in the first embodiment, if the threshold voltage value th of the field effect transistor is positive, it is assumed that a voltage (0.5th) of the field-effect transistor is applied to the control line for reading. In addition, although the conventional FeFET and the FeFETs of Embodiments 1 and 2 write a voltage of Vg = ± 1 5 V to the control electrode in the writing system, this embodiment 3 is writing

第23頁 490859 案號 90105477 _ ' 五、發明說明(ϊ^ &quot; ' — ----- 入電壓設定為Vg = ± 3V。 於本實施例3,圖5八係顯示施 = 貝枓0」寫入之動作狀態圖;圖6係顯f ~ 3V後之 0.5V後之資料讀出之動作狀態圖。•不知加Vg = 於圖 5A、圖 5B、圖6 上,61、63、69 7“糸顯示增加控制電極電壓時之鐵電、:、75、 線,62、64、68、7〇係顯示 =性之曲 ,特性之曲線,各自的初期偏極化值==鐵,體 ;料於圖05'圖5β上,65為資料「1」寫入動作66同為 2 J」&quot;作,67為保持動作,於圖6上之72#顯 不項出動作時之FET的Qi對(ν 口上之72係顯 84係表示動作點。 、f之曲線,77〜 針對由一個初期狀態而寫入 用,加以說明。初期狀態係指於 」的二^ 保持之資料與相同之資料的寫;△貝枓狀恶。比較將所 始之眘m宜a ,Bll、、卄的寫動作,若從該初期狀態開 二可以H A :: ΐ同的施加電壓所得到的偏極化值較 T U说疋因為嚴可的動作條件所致。 平方:ί二正電壓於初期狀態為偏極化值0庫倫/ 二分之鐵電體’如同Pdfm特性61。當施力七=3v :0· 對(Vg ~ Vf)特性為曲線65,pd對”特性61 2對二V;?)特性65之交點A為資料Π」寫入動 ㈣n ^控制電極設定為浮置,如前述之m的Page 23 490859 Case No. 90105477 _ 'V. Description of the invention (ϊ ^ &quot;' — ----- The input voltage is set to Vg = ± 3V. In this embodiment 3, Fig. 5 shows the eighth series of application = 枓 0 "Writing operation state diagram; Figure 6 shows the state of data reading after 0.5V after f ~ 3V. • I don't know if Vg = 61, 63, 69 on Figure 5A, Figure 5B, and Figure 6 7 "糸 shows the ferroelectricity: ,, 75, line when the control electrode voltage is increased, 62, 64, 68, 70 series display = curve of characteristics, characteristic curve, each initial polarization value = = iron, body It is expected that in Fig. 05 'and Fig. 5β, 65 is the data "1" write operation 66 is the same as 2 J "&quot; operation, 67 is the holding operation, and 72 # in Fig. 6 shows the FET when it is not in operation Qi indicates that the 72 points on the ν mouth and 84 points indicate the operating points. The curve of f, 77 ~ will be described for writing from an initial state. The initial state refers to the data held by the two ^ The writing of the data; △ beauty-like evil. Compare the initial cautious action of a, Bll, and 开. If you start from this initial state, you can use HA :: different polarities obtained by applying different voltages. Compared with TU, the value is due to the strict operating conditions. Square: ί Two positive voltages in the initial state have a polarization value of 0 Coulombs / half of a ferroelectric 'like Pdfm characteristic 61. When the force 7 = 3v: 0 · Pair (Vg ~ Vf) characteristic is curve 65, pd versus "characteristic 61 2 versus two V;?) Intersection point A of characteristic 65 is data Π" Write motion ㈣ n ^ The control electrode is set to float, as described above m of

IMI II H-yuojy 修正IMI II H-yuojy correction

號 90105477 五、發明說明(20) U^Vg 一Vf)特性67,趨近於Vg〜〇V附近,則谁入伴 於,&amp;丨1 1」保持動作點係從寫入動作點門 始減少電壓時的Pd對”特性62 乍/開No. 90105477 V. Description of the invention (20) U ^ Vg-Vf) Characteristic 67, which is close to Vg ~ 〇V, then who is involved, &amp; 丨 1 1 "Keeping the action point starts from the write action point gate Pd pair at reduced voltage "characteristic 62

Vf)特性67之交。伴拉翻从机g VB卞的Ql對(k 一 -0 42V。,Ϊ /動作點之動作電壓為h = 設定的電;Γ 設定為迅速地㈣對…1)特性 於鐵電體以初期值將資料「 官人夕味 對Vf特性63與於Vg = 3 」寫m兄下,Pd 點C為資料「1 f宜入包从對(g — Vf)特性65之交 電極電壓時之鐵電體特性“二。’特咸少控制 :0. 汲 0 j 「1如二來從二個初期狀態開始,於已寫入資料 1」保持之FET漂移閘極電極 、 42V&amp;〇.39V,成為Vi &lt;柽電壓為Vl =。— Vx 極電流並不流動: 。亦即,FET為關閉狀態 接著’針對從二個初期 寫入之動作,利用㈣:月狀恕開始’針對將資料 、 〜用圖5 B加以說明。 初期狀態下,你‘名+ ^ 分之鐵電體時,則成壓於偏極化值〇庫倫/平方公 之FET的Qi對(v v、f 〇對”特性68。施加Vg = -3V時 Qi對(Vg - Vn特性為曲線66,?(1對”特性68與Vf) Turn of characteristic 67. The Ql pair with the pull-over slave g VB 卞 (k-0 42V., The operating voltage of Ϊ / action point is h = set electricity; Γ is set to quickly ㈣ pair ... 1) Characteristics in the early stage of ferroelectric Based on the data "Official people's taste of Vf characteristics 63 and Vg = 3", write "m", Pd point C is the data "1 f should be included in the pair of (g-Vf) characteristics 65 at the cross electrode voltage of iron Electrical characteristics "two. 'Extremely small control: 0. Drain 0 j "1 as two come from two initial states, in the written data 1" FET drift gate electrode, 42V &amp; 0.39V, becomes Vi &lt; 柽 voltage Vl =. — Vx pole current does not flow:. In other words, the FET is in the off state. Next, "the operation of writing from the two initial stages will be described using" ㈣: month shape forgiveness ". The data will be described with reference to Fig. 5B. In the initial state, when you're ferroelectric with the name + ^, it becomes a Qi pair (vv, f 0 pair) characteristic of the FET with a bias value of 0 Coulomb / square centimeter. When Vg = -3V is applied Qi pair (Vg-Vn characteristics are curve 66,? (1 pair) characteristics 68 and

於此,將控制電極浮置或施加vg = ov,」則㈡:Here, float the control electrode or apply vg = ov, "then ㈡:

案號 90105477 五、發明說明(21) 動作。此資料「0」保持動作點佐&gt; 増加電壓時的Pd對Vf特性fiq命v ‘ .,、從寫入動作點開始, 特性67之交點F。保持動作點、g %〇V時的…對(Vg — Vf) 鐵電體為初期值之資料「作電壓為Vx = 0.4V。 對Vf特性70與Vg = -3V時之〇彳料」μ已寫入之情況下,Pd 點G為資料「〇」寫入動 :&quot;Ί Vf)特性66之交 電極電壓時之鐵電體特性盥,該狀態開始,增加控制 交點Η成為保持動作點。持對(V§ — V〇特性67之 38V。 ,、符動作點之動作電壓為Vx = 0· 如此來’從二個初期狀能門 「〇」保持之FET潭移n ^心開始,於已寫入資料 一 0.40V 及-0.38V ’ 成為Vi 電£為^ = Vg - VX = 態,汲極電流並不流動。Vth。亦即,FET為關閉狀 針對從二個初期狀能 出之動作,利用圖6加:說開明始。?十對將資料「1」已讀 電壓於控制電極上。 本動作係施加V g = 〇 · 3 V之 從偏極化值〇庫检/ 期狀態開始,已寫入的資千;广與資肖「〇」之二個初 78,從各自的保持動、 」保持動作點為77、 對Vf特性73、74盥施Λ v幵始’增加電壓時的鐵電體之1^Case No. 90105477 V. Description of Invention (21) Action. This data "0" holds the operating point and &gt; The Pd vs. Vf characteristic fiq at the time of voltage application is given by v '., Starting from the writing operating point, at the intersection F of characteristic 67. When the operating point is maintained, g% 〇V ... For (Vg — Vf) Ferroelectrics is the initial value of the data "Operating voltage is Vx = 0.4V. For Vf characteristics 70 and Vg = -3V ○ material" μ When it has been written, the Pd point G is the data "0". The motion is: "quote Vf) The ferroelectric characteristics at the cross electrode voltage of characteristic 66. When this state starts, increase the control intersection Η to become the holding operation point. . Holding pair (V§ — 38V of V0 characteristic 67., The operating voltage of the run point is Vx = 0. So 'start from the shift of the FET n n centered by the two initial energy gates "0", at Have written data-0.40V and -0.38V 'become Vi electricity. It is ^ = Vg-VX = state, and the drain current does not flow. Vth. That is, the FET is turned off for two initial states. Action, use Figure 6 to add: Let ’s say enlightenment. Ten pairs of data “1” read the voltage on the control electrode. This action is to apply V g = 〇 · 3 V from the polarization value 〇 library inspection / period state At the beginning, the written information has been written; the first two 78 of the "0" of Guang and Zixiao, from the respective holding action, the holding action point is 77, and the voltage Vf characteristics 73 and 74 are applied to increase the voltage. 1 of the time ferroelectric

Vf)特性72之交點8tH3V時之FET _對(1 為Vx = -〇. 3V及-〇. 28v。Vi靖出動作點。此讀出動作電壓 =,由此較臨限電壓值大:: :x = 〇.6V及0.5”成 導通。因此,資料π 、/ Vth)之故,FeFET成為 、 '被正確地讀出。Vf) FET _ pair (1 is Vx = -0.3V and -0.28v at the intersection of characteristic 72. Vi is the operating point. This reads the operating voltage =, which is greater than the threshold voltage value :: : x = 0.6 V and 0.5 "are turned on. Therefore, the data π, / Vth), the FeFET will be read correctly.

JJ

第26頁 490859 案號 9〇ιηΜ77 五、發明說明(22) 曰 一修正 接著,利用圖6說明針對從二個初 料「〇」寫入之動作。 ’ 4開始,將資 從偏極化值0庫倫/平方公分與資料Γ1 ^狀態開始,已寫入的資料「〇」 」之二個初 8〇,從各自的保持動作點#,增 壓乍點為79、 Π特性75、76與施加Vg = 〇3 J的,電體之Pd對 特性…點83、84為讀出動作;广^ &quot;.m及。·44ν„νί = Vg _ νχ = _〇.二出動作電壓為Vx 立,因此較臨限電壓值小(Vi〈 .14V成 因此,可正確的讀出資料「〇」。)’ FeFET成為關閉。 如此一來,於本發明之實施例3,^料 〇」之保持動作上,FeFET均為關 」” 加0.3V而能進行正確的讀出』作1。“:土由於施 使於汲極電極與位元線間曰 用,即 的,也能組成m j 選擇電晶體’習知是必須 的實施例。 〜70歹'。於圖7係顯示有關於其具體 於圖7,由89之鐵電體電容、9〇之阳 FeFET。9 1表示閘極選擇電晶體 、'’ 夹干篦彳仝- &amp; / 體92表不位兀線(BL)、86 ΐίί « )、85表示第2字元線(WL2)、87表示 源極線(SL)、88表示記憶單元的基本單位。 數字ΖΛ2Γ列之記憶*元陣列,附加於各構成要素之 表之仃的編號或是列的編號。例如,8 6 -1、 列之礼別表示第1行之WU、WL2、SL ,92 — 1表示第1 歹J之BL、88-11表示第】行第!列之記憶單元。 第27頁 490859 案號 90105477 __- 五、發明說明(23) 於如此組成之記憶單元陣列,針對位於任意位址之記 憶單元陣列’藉由適用於上述之驅動方法,做為具有機能 的不揮發性兄憶體。於圖8中以表格歸納該驅動方法。 例如’將已寫入圖7的§己憶單元陣列之第1行第1列 (圖7的88-11)的讀出資料時,wli (圖7的86一u設定為 5V而閘極選擇電晶體設定為導通,施加〇· 3V於乳?(圖7'的 85-1)。此時,該記憶單元之FeFET的控制電極成為”=、 0.3V ’若施加0.3V於BL (圖7的92-1)、施加(^於讥(圖7 ,87-1) ’則汲極電流會從Bl流向Sl。於BL的終端上連接 ,出放大器,若藉由汲極電流以檢測電壓降低,根據保持 資料而改變汲極電流值之故,則能進行正確的讀出動作。 此時,同樣地連接於BL的第2行第!列之記憶單元(圖7的 88-21),雖然處於保持動作狀態下,如前所述,因為汲極 ?流並不流動,對於第&quot;于第!列之記憶單元讀出動作並不 又衫響。因此,能組成省略汲極電極· BL間之選擇電晶體 的記憶單元。 還有,於本發明之實施例3,藉由寫入電壓之調整, 將保持動作中的FeFET設定為關閉,除此種方法之外,例 如藉由FET之通道摻雜而控制臨限電壓值,也可以將 ί I! t,eFET设定為關Μ。同時,調整構成FeFET之鐵電 FET /*、比誘電率、阻抗電壓等之材料特性,或是調整 二之閘極絕緣膜的膜厚、比誘電率、阻抗電壓等之材料 # ώ : ^將㈣動作中的FeFET設定為關閉4適用於 精由该之荨臨限電壓值控制的方法,寫人電壓之設定電壓 490859 ___縫90105477_年月曰 修正___ 五、發明說明(24) 值,於施加電壓後之無施加電壓狀態下,上述場效應電晶 體處於成為非導通狀態之電壓值的範圍内,並能具有同樣 的效果。P.26 490859 Case No. 9〇ηη77 77 V. Description of the Invention (22) Revision 1 Next, the operation for writing from two preliminary data “0” will be described using FIG. 6. From the beginning of 4, starting from the polarization value of 0 Coulomb / cm 2 and the data Γ1 ^ state, the written data "0" two first 80, from the respective holding action point #, pressurization The points are 79, Π characteristics 75, 76 and the applied Vg = 〇3 J, the Pd pair characteristics of the electrical body ... points 83, 84 are read-out actions; wide ^ &quot; .m and. · 44ν „νί = Vg _ νχ = _〇. The output voltage of the second output is Vx, so it is smaller than the threshold voltage (Vi <.14V, so the data“ 〇 ”can be read correctly.) 'FeFET is turned off . In this way, in the holding operation of the third embodiment of the present invention, the "FeFETs are all off", and "0.3V is added to enable correct reading." The electrode and the bit line are used immediately, that is, they can also form an mj-selecting transistor, which is a necessary embodiment. ~ 70 歹 '. Fig. 7 shows details of the ferroelectric capacitor 89 and the 90 FeFET, which are specific to Fig. 7. 9 1 indicates the gate selection transistor, '' interfering with-&amp; / body 92 (BL), 86 ΐίί «), 85 indicates the second character line (WL2), 87 indicates the source The polar lines (SL), 88 represent the basic units of the memory unit. A memory * element array of the number ZΛ2Γ column is appended to the number of the element or the number of the column of the table of each constituent element. For example, 8 6 -1, column ritual category means WU, WL2, SL in the first row, 92 — 1 means BL in the first 歹 J, 88-11 means row]! List of memory cells. Page 27 490859 Case No. 90105477 __- V. Description of the invention (23) Based on the memory cell array composed in this way, the memory cell array located at an arbitrary address is used as the non-volatile functional device by applying the driving method described above. Sex brother recalls body. The driving method is summarized in a table in FIG. 8. For example, when reading the data that has been written into the 1st row and 1st column (88-11 in FIG. 7) of the § memory cell array of FIG. 7, wli (86-u in FIG. 7 is set to 5V and the gate is selected The transistor is set to be on, and 0.3 V is applied to the breast? (85-1 in Fig. 7 '). At this time, the control electrode of the FeFET of the memory cell becomes "=, 0.3 V' if 0.3 V is applied to BL (Fig. 7). 92-1), apply (^ to 讥 (Figure 7, 87-1) ', then the drain current will flow from Bl to Sl. Connect to the terminal of BL, and output the amplifier. If the drain current is used to detect the voltage drop If the drain current value is changed according to the holding data, a correct reading operation can be performed. At this time, it is similarly connected to the memory cell of the second row and the second row of BL (88-21 in FIG. 7), although In the state of holding action, as mentioned above, because the drain current does not flow, the reading operation of the memory cells in the "!" Column does not cause any noise. Therefore, the drain electrode can be omitted. BL The memory cell of the transistor is selected in between. In addition, in Embodiment 3 of the present invention, the FeFET in the holding operation is set to off by adjusting the write voltage, except that In addition to this method, for example, by controlling the threshold voltage value by channel doping of the FET, it is also possible to set ί! T, eFET to OFF M. At the same time, adjust the ferroelectric FET / *, specific induction power constituting the FeFET. Material characteristics such as rate, impedance voltage, etc., or materials that adjust the film thickness, specific induction, impedance voltage, etc. of the gate insulation film # 2: ^ Set the FeFET during the operation to off 4 The method of controlling the threshold voltage value of the net, write the set voltage of the human voltage 490859 ___9090477_year, month, month, month, month, month, month, month, month, month, month, month, month, month, month, month, month, month, month, month, month, month, month, month, month, month, month, month, month, month, month, month, month, month, month, month, month, month, month, month, month, month, month, month, month, month, month, month, month, month, month, month, month, month, month, month, month, month, month, month, month, month, month, month, month, month, month, month, month, month, month, month, month, month, month, month, month, month, month, month for the month, and the day of the month. The effect transistor is in the range of the voltage value in the non-conducting state, and can have the same effect.

還有,於本說明書中,已針對MFMIS型FeFET加以描 述,MFIS型FeFET之情形也可以經由同樣的考量而設定動 作電壓(寫入電壓及讀出電壓),該構造之動作電壓設定 法也包含於申請專利範圍内。亦即,考量本發明中之漂移 閘極電極電壓為鐵電體與絕緣膜之界面上的電位,算出對 於該界面電位與通道所感應的電荷(本說明書中之而 與,道成為導通之電壓(本說明書中之vth)等之關係而 設定電壓,可以得到與本發明同樣的效果。Also, in this specification, the MFMIS-type FeFET has been described. In the case of the MFIS-type FeFET, the operating voltage (writing voltage and reading voltage) can be set through the same considerations. The operating voltage setting method of this structure also includes Within the scope of patent application. That is, considering the drift gate electrode voltage in the present invention as the potential at the interface between the ferroelectric and the insulating film, the charge induced to the interface potential and the channel is calculated (and in this specification, the voltage at which the channel becomes conductive) Setting the voltage according to the relationship (vth in this specification) and the like can obtain the same effect as the present invention.

第29頁 490859 -- 素號.90105477__年月日__ 圖式簡單說明 圖1係顯示資料「1」與資料「〇」之Ids比(on / of f比)與Vg之關係圖。 圖2係有關於本發明實施例1之半導體記憶裝置之動作 狀態圖。 圖3A〜C係有關於本發明實施例2之半導體記憶裝置之 動作狀態圖。 圖4A、圖4B係有關於本發明實施例2之半導體記憶裝 置之動作狀態圖。 圖5 A、圖5 B係有關於本發明實施例3之半導體記憶裝 置之動作狀態圖。 圖6係有關於本發明實施例3之半導體記憶裝置之動作 狀態圖。 圖7係有關於本發明實施例3之半導體記憶裝置之陣列 圖。 圖8係顯示有關於本發明實施例3之半導體記憶裝置之 驅動電壓表格。 歩署表示有關於本發明實施例及習知之半導體記憶 装置構成之剖面構造圖。 f置==有關於本發明實施例及習知之半導體記憶 衮置構成之等價迴路圖。 態圖圖1&quot;、圖m係為習知之半導體記憶装置之動作狀 圖。隨、關係為習知之半導體記憶裝置之動作狀態Page 29 490859-Prime No. 90105477__ 年月 日 __ Brief description of the diagram Figure 1 shows the relationship between the Ids ratio (on / of f ratio) and Vg of the data "1" and the data "0". Fig. 2 is a diagram showing an operation state of the semiconductor memory device according to the first embodiment of the present invention. 3A to 3C are diagrams showing operation states of the semiconductor memory device according to the second embodiment of the present invention. Figs. 4A and 4B are diagrams showing operation states of the semiconductor memory device according to the second embodiment of the present invention. 5A and 5B are diagrams showing the operation state of the semiconductor memory device according to the third embodiment of the present invention. Fig. 6 is an operation state diagram of the semiconductor memory device according to the third embodiment of the present invention. Fig. 7 is an array diagram of a semiconductor memory device according to a third embodiment of the present invention. Fig. 8 shows a driving voltage table of a semiconductor memory device according to a third embodiment of the present invention. The Agency has shown a cross-sectional structure diagram of the structure of a conventional semiconductor memory device according to an embodiment of the present invention. f set == There is an equivalent circuit diagram of the embodiment of the present invention and the conventional semiconductor memory configuration. The state diagrams Figure 1 and Figure m are operation diagrams of conventional semiconductor memory devices. Follow-up relationship is the operating state of a conventional semiconductor memory device

第30頁 490859 _案號90105477_年月曰 修正_ 圖式簡單說明 圖1 3係為習知之半導體記憶裝置之動作狀態圖。 [符號說明] 85-1 :第1行之字元線WL2 85- 2 :第2行之字元線WL2 86- 1 :第1行之字元線WL1Page 30 490859 _Case No. 90105477_ Year Month Revision _ Brief Description of Drawings Figure 13 is a diagram of the operating state of a conventional semiconductor memory device. [Explanation of symbols] 85-1: Character line WL2 in the first row 85-2: Character line WL2 in the second row 86-1: Character line WL1 in the first row

86- 2 ··第2行之字元線WL1 8 7-1 :第1行之源極線SL86- 2 · Character line WL1 in the second line WL 1 8 7-1: Source line SL in the first line

87- 2 :第2行之源極線SL87- 2: Source line SL in line 2

88- 11 ··第1行第1列之記憶單元 88-12 :第1行第2列之記憶單元 88-21 :第2行第1列之記憶單元 8 8-22 :第2行第2列之記憶單元 89 :鐵電體電容88-11: Memory cell in row 1 and column 88-12: Memory cell in row 1 and 2 88-21: Memory cell in row 2 and 1 8-22: 2nd row and 2 Memory Unit 89: Ferroelectric Capacitor

90 : FET 91 :閘極選擇電晶體90: FET 91: Gate selection transistor

9 2-1 :第1列之位元線BL 9 2-2 :第2列之位元線BL9 2-1: Bit line BL in the first row 9 2-2: Bit line BL in the second row

31 基 板 32 汲 極 電 極 之 高 濃 度N 型 植入 區域 33 源 極 電 極 之 高 濃 度N 型 植入 區域 34 介 電 體 35 漂 移 閘 極 電 壓 36 鐵 電 體31 Substrate 32 High-concentration N-type implanted area of drain electrode 33 High-concentration N-type implanted area of source electrode 34 Dielectric 35 Drift gate voltage 36 Ferroelectric

第31頁 490859 _案號90105477_年月日_修正 圖式簡單說明 37 控 制 電 極 41 鐵 電 體 電 容 42 FET 43 控 制 電 極 44 源 極 電 極 45 汲 極 電 極 Vd 汲 極 電 極 之 電 壓 Vf 電 極 間 電 壓 Vg 控 制 閘 極 電 壓 Vi 漂 移 閘 極 之 電 壓 Vs 源 極 電 極 之 電 壓Page 31 490859 _Case No. 90105477_Year_Month_Revision of the schematic diagram 37 Control electrode 41 Ferroelectric capacitor 42 FET 43 Control electrode 44 Source electrode 45 Drain electrode Vd Drain electrode voltage Vf Voltage between electrodes Vg Control gate voltage Vi Drift gate voltage Vs Source electrode voltage

第32頁Page 32

Claims (1)

------— 六、申請專利範圍 1β 一種半導體記憶裝置之資料4山 電極與強介電體所構成的電容之貝AV7*,由成對的 成; 〃逼%效應型電晶體所組 上述電容之一側電極連接於ν 閑極電極,或是上述電容之一側電=效應型電晶體之 型電晶體之閘極電極; 側電極兼作為上述電場效應 ,上述電容之另一侧電極做為控制電極, 施加電 反於上述控制電極以改變上 9 上沭® π 4、5$ ;丨電體之偏極化,而改變 應型電晶體之通道阻抗,依上述通道阻抗之高 低而表不成二值資料; 其特徵為: =上述電場效應型電晶體之臨界電壓值為正,則施加 正電壓於上述控制電極; 鱼上述電場效應型電晶體之臨界電壓值為負,則施加 負電壓於上述控制電極。 2·如申請專利範圍第1項之半導體記憶裝置之資 出方法,其中: 十5貝 施加於上述控制電極的電壓值為: 當上述電場效應型電晶體之臨界電壓值為正的情形, 於上述電容之殘留偏極化值為〇 C/cm2之狀態下,將電極 間電麼由0V開始增加使偏極化值改變時,臨界電壓施加於 上述電場效應型電晶體之閘極電極時,將產生相當於對通 道所感應產生之電荷的偏極化值之電極間電壓值,與上述 第33頁 49〇859 力、申請專利範圍 電場效應型電晶體之臨界電壓值相加,以所得到之電壓值 為中心 ±20%以内之電壓值; 當上述電場效應型電晶體之臨界電壓值為負的情形, 於上述電容之殘留偏極化值為0 C / c m2之狀態下,將電才蛋 間電壓由0 V開始減少使偏極化值改變時,臨界電壓施加於 上述電場效應型電晶體之閘極電極時,將產生相當於對通 道所感應產生之電荷的偏極化值之電極間電壓值,與上述 電場效應型電晶體之臨界電壓值相加,以所得到之電壓值 為中心± 20%以内之電壓值。 3. 電極與 成; 上 閘極電 型電晶 將 壓於上 上述電 低而表 其 於 施加電 狀態之 一種半導體記憶裝置之資料寫入方法,由成對的 強介電體所構成的電容,與電場效應型電晶體所組 f電容之一側電極連接於上述電場效應型電晶體之 =是上述電容之一侧電極兼作為上述電場= 體之閘極電極; i ί 2 ϊ之另一侧電極做為控制電極,藉由施加電 述rr電體之偏極化,而改變 示成-ϊί通逞阻抗,依上述通道阻抗之高 下战一值資料; 特徵為: 壓極所施加之電壓值處於施加電壓後之無 範圍内。上述電場效應型電晶體係處於非導通-------- 6. Scope of patent application 1β Information of a semiconductor memory device 4 Capacitor AV7 * composed of a mountain electrode and a ferroelectric body, which are formed in pairs; One side electrode of the above capacitors is connected to the ν idle electrode, or one of the above capacitors is a gate electrode of a transistor of the effect type transistor; the side electrode also serves as the electric field effect, and the other side of the capacitor The electrode is used as the control electrode. Applying electricity to the above control electrode to change the upper 9 沭 ® π 4,5 $; 丨 the polarization of the electric body, and change the channel impedance of the application transistor, according to the level of the above channel impedance The table is not a binary data. Its characteristics are: = If the critical voltage value of the electric field effect transistor is positive, a positive voltage is applied to the control electrode; If the critical voltage value of the electric field effect transistor is negative, the voltage is applied. A negative voltage is applied to the control electrode. 2. If the method for funding a semiconductor memory device according to item 1 of the scope of the patent application, wherein: the voltage value applied to the control electrode is: When the threshold voltage value of the electric field effect transistor is positive, in When the residual polarization value of the capacitor is 0C / cm2, when the voltage between electrodes is increased from 0V to change the polarization value, when a critical voltage is applied to the gate electrode of the electric field effect transistor, The inter-electrode voltage value that generates a polarization value corresponding to the charge induced by the channel is added to the above-mentioned 49 859 force on page 33, the critical voltage value of the field-effect transistor in the patent application range, to obtain The voltage value is within ± 20% of the center. When the critical voltage value of the electric field effect transistor is negative, the voltage of the capacitor is 0 C / c m2. When the voltage between eggs starts to decrease from 0 V and the polarization value is changed, when the critical voltage is applied to the gate electrode of the above-mentioned electric field effect transistor, a charge equivalent to the charge induced in the channel will be generated. The voltage value between the electrodes of the polarization value is added to the above-mentioned critical voltage value of the electric field effect transistor, and the voltage value within the center of the obtained voltage value is within ± 20%. 3. Electrodes and formation; the data writing method of a semiconductor memory device in which the upper gate electric transistor will be pressed above the above low level, indicating that it is in an applied state, and a capacitor composed of a pair of ferroelectric bodies , One side electrode connected to the f capacitor of the electric field effect transistor is connected to the electric field effect transistor, which is one side electrode of the capacitor and serves as the gate electrode of the electric field body; i ί 2 ϊ the other The side electrode is used as the control electrode. By applying the polarization of the rr electric body, the impedance is changed to-ϊ 逞 through 依 impedance, according to the above-mentioned channel impedance and the value of the data; Features: The voltage value is within the range after the voltage is applied. The above electric field effect type transistor system is non-conducting 第34頁 490859 六、申請專利範圍 4· 一種半導體記憶裝置之驅動方法,在藉由申請專 範圍第3項之半導體記憶裴置之資料寫入方法,進行資 料寫入之後,施加重新設定的電壓於上述控制電極。 5· 一種半導體記憶裝置之資料讀出方法,用以將依 ,申凊f利範圍第3項之半導體記憶裝置之資料寫入方法施 仃過貧料寫入的半導體記憶裝置之資料予以讀出,其特徵 為: 士於上述電場效應型電晶體之臨界電壓值為正的情形 時,施加正電壓於上述控制電極,並且於上述電場效應型 電晶體施加低於臨界電壓值的電壓; 若上述電場效應型電晶體之臨界電壓值為負,則施加 負f壓於上述控制電極,並且於上述電場效應型電晶體施 加高於臨界電壓值的電壓。 6 ·如申請專利範圍第5項之半導體記憶裝置之資料讀 出方法,其中: 、 ^ 由上述成對的電極與強介電體所構成的電容,與上述 電場效應型電晶體所構成的半導體記憶裝置排列成複數個 矩陣狀; 上述電場效應型電晶體之汲極電極連接於位元線、源 極電極連接於源極線、第2電場效應型電晶體之源極電極 連接於控制電極;Page 34 490859 VI. Application for Patent Scope 4. A method for driving a semiconductor memory device. After applying the data writing method for the semiconductor memory device in the third item of the application scope, after the data is written, a reset voltage is applied. On the control electrode. 5. · A method for reading data from a semiconductor memory device, for reading the data from the semiconductor memory device written in the lean material according to the third item of the claim range. , Characterized in that: when the critical voltage value of the electric field effect transistor is positive, a positive voltage is applied to the control electrode, and a voltage lower than the critical voltage value is applied to the electric field effect transistor; if the above If the critical voltage value of the electric field effect transistor is negative, a negative f voltage is applied to the control electrode, and a voltage higher than the critical voltage value is applied to the electric field effect transistor. 6 · The method for reading data from a semiconductor memory device according to item 5 of the scope of patent application, wherein: ^ a capacitor composed of the pair of electrodes and a ferroelectric body, and a semiconductor composed of the electric field effect transistor described above The memory devices are arranged in a plurality of matrices; the drain electrode of the electric field effect transistor is connected to the bit line, the source electrode is connected to the source line, and the source electrode of the second electric field effect transistor is connected to the control electrode; 第35頁 490859Page 490 490859 第36頁Page 36
TW090105477A 2000-03-09 2001-03-07 Methods of reading and writing data from/on semiconductor memory device, and method for driving the device TW490859B (en)

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
JP2000064925A JP2001256774A (en) 2000-03-09 2000-03-09 Data read-out method for semiconductor memory, data write-in method, and driving method

Publications (1)

Publication Number Publication Date
TW490859B true TW490859B (en) 2002-06-11

Family

ID=18584545

Family Applications (1)

Application Number Title Priority Date Filing Date
TW090105477A TW490859B (en) 2000-03-09 2001-03-07 Methods of reading and writing data from/on semiconductor memory device, and method for driving the device

Country Status (6)

Country Link
US (1) US6438021B2 (en)
EP (1) EP1132921A1 (en)
JP (1) JP2001256774A (en)
KR (1) KR20010100811A (en)
CN (1) CN1245721C (en)
TW (1) TW490859B (en)

Families Citing this family (15)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
JP3749851B2 (en) * 2001-10-25 2006-03-01 株式会社東芝 Ferroelectric semiconductor memory
AU2002357606A1 (en) * 2001-12-20 2003-07-09 Matsushita Electric Industrial Co., Ltd. Potential generating circuit, potential generating apparatus, semiconductor device using the same, and driving method thereof
JP3520871B2 (en) 2002-01-31 2004-04-19 ソニー株式会社 Magnetizing method and magnetizing device
CN1303692C (en) * 2002-09-04 2007-03-07 松下电器产业株式会社 Semiconductor memory device, method for fabricating the same, and method for driving the same
JP2004139655A (en) * 2002-10-17 2004-05-13 Matsushita Electric Ind Co Ltd Semiconductor storage device and electronic equipment provided with the same
US7016217B2 (en) * 2003-03-07 2006-03-21 Sanyo Electric Co., Ltd. Memory
JP4452631B2 (en) 2005-01-21 2010-04-21 パトレネラ キャピタル リミテッド, エルエルシー memory
KR100695139B1 (en) * 2005-02-07 2007-03-14 삼성전자주식회사 Ferroelectric recording medium, and writing method of the same
DE102006010979B3 (en) * 2006-03-09 2007-04-12 Infineon Technologies Flash Gmbh & Co. Kg Read voltage adjusting method for e.g. nitride read only memory, involves utilizing adjusted read voltage or read voltage within adjusted voltage range in normal mode of memory to read data from memory cells
KR101642819B1 (en) * 2009-08-31 2016-07-26 삼성전자주식회사 Nonvolatile memory device, driving method thereof and memory system having the same
US20160005749A1 (en) * 2014-07-01 2016-01-07 Qualcomm Incorporated Series ferroelectric negative capacitor for multiple time programmable (mtp) devices
US10033383B1 (en) * 2017-03-20 2018-07-24 Globalfoundries Inc. Programmable logic elements and methods of operating the same
US11502103B2 (en) * 2018-08-28 2022-11-15 Intel Corporation Memory cell with a ferroelectric capacitor integrated with a transtor gate
US11980037B2 (en) 2020-06-19 2024-05-07 Intel Corporation Memory cells with ferroelectric capacitors separate from transistor gate stacks
US11764255B2 (en) * 2021-04-28 2023-09-19 National Central University Memory circuit, memory device and operation method thereof

Family Cites Families (13)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
JP3153606B2 (en) 1992-01-27 2001-04-09 ローム株式会社 Nonvolatile storage element, nonvolatile storage device using the same, and method of driving nonvolatile storage device
JP3176114B2 (en) * 1992-01-27 2001-06-11 ローム株式会社 Nonvolatile storage element, nonvolatile storage device using the same, and method of driving nonvolatile storage device
JP2921812B2 (en) * 1992-12-24 1999-07-19 シャープ株式会社 Nonvolatile semiconductor memory device
US5432731A (en) * 1993-03-08 1995-07-11 Motorola, Inc. Ferroelectric memory cell and method of sensing and writing the polarization state thereof
JPH0745794A (en) * 1993-07-26 1995-02-14 Olympus Optical Co Ltd Drive method for ferroelectric memory
US5753946A (en) * 1995-02-22 1998-05-19 Sony Corporation Ferroelectric memory
KR100218275B1 (en) * 1997-05-09 1999-09-01 윤종용 Ferroelectric memory device with bulk-type one transistor structure
JPH1117123A (en) * 1997-06-23 1999-01-22 Rohm Co Ltd Non-volatile memory element
US6067244A (en) * 1997-10-14 2000-05-23 Yale University Ferroelectric dynamic random access memory
US6191441B1 (en) * 1997-10-28 2001-02-20 Fujitsu Limited Ferroelectric memory device and its drive method
JP3780713B2 (en) * 1998-08-25 2006-05-31 富士通株式会社 Ferroelectric memory, ferroelectric memory manufacturing method, and ferroelectric memory testing method
DE19840824C1 (en) * 1998-09-07 1999-10-21 Siemens Ag Ferroelectric transistor especially for a non-volatile memory cell
US6233169B1 (en) * 1998-11-06 2001-05-15 Rohm Co., Ltd. Signal storing circuit semiconductor device, gate array and IC-card

Also Published As

Publication number Publication date
KR20010100811A (en) 2001-11-14
CN1313603A (en) 2001-09-19
US20010050367A1 (en) 2001-12-13
CN1245721C (en) 2006-03-15
JP2001256774A (en) 2001-09-21
US6438021B2 (en) 2002-08-20
EP1132921A1 (en) 2001-09-12

Similar Documents

Publication Publication Date Title
TW490859B (en) Methods of reading and writing data from/on semiconductor memory device, and method for driving the device
JP2965415B2 (en) Semiconductor storage device
TWI338302B (en) Method of read, programming and erasing a p-channel be-sonos nand flash memory
TW548825B (en) Semiconductor device
US6898105B2 (en) Ferroelectric non-volatile memory device having integral capacitor and gate electrode, and driving method of a ferroelectric non-volatile memory device
JP3505758B2 (en) Non-volatile semiconductor memory
US7339821B2 (en) Dual-gate nonvolatile memory and method of program inhibition
JPH0745794A (en) Drive method for ferroelectric memory
US11094820B2 (en) Mobile ferroelectric single domain wall implementation of a symmetric resistive processing unit
US20110248324A1 (en) Dual-gate non-volatile ferroelectric memory
JP3906177B2 (en) Nonvolatile semiconductor memory device
JP2004039965A (en) Nonvolatile semiconductor storage device
TW201039437A (en) Bandgap engineered charge trapping memory in two-transistor NOR architecture
TW413818B (en) Programming and erasing method for a split gate flash EEPROM
TW201131568A (en) Novel punch-through free program scheme for NT-string flash design
KR20010030545A (en) Non-volatile memory
US8675381B2 (en) Transistor having an adjustable gate resistance and semiconductor device comprising the same
TWI244165B (en) Single bit nonvolatile memory cell and methods for programming and erasing thereof
US20010019497A1 (en) Semiconductor memory device, method for driving the same and method for fabricating the same
JP2008153479A (en) Semiconductor device with ferroelectric field-effect transistor, and semiconductor integrated circuit device using the same
US6812083B2 (en) Fabrication method for non-volatile memory
JP2011023705A (en) Nonvolatile semiconductor memory device
US6760246B1 (en) Method of writing ferroelectric field effect transistor
JPS6180866A (en) Non-volatile semiconductor memory cell
TW492180B (en) Semiconductor memory and method for driving the same

Legal Events

Date Code Title Description
GD4A Issue of patent certificate for granted invention patent
MM4A Annulment or lapse of patent due to non-payment of fees